Detecting object distance using Depth Camera and SSD-Mobilenet

Follow along: Detecting object distance using Depth Camera and SSD-Mobilenet

The program launching process along with parameter settings are all simplified and set up on the Jupyter Notebook Environment.
  • Open the 01_07_Detect_Object_and_Calculate_Distance.ipynb Jupyter Notebook.
  • This code is output by applying numpy and matplotlib.
  • This code uses a pre-trained model.
(The Jetson Board used for these examples are => Jetson Nano)

  • 01_07_Detect_Object_and_Calculate_Distance.ipynb

  • Running the cell code.
    Ctrl + Enter

  • Load the modules needed to run your code.

import cv2
import numpy as np
import matplotlib.pyplot as plt
import pyrealsense2 as rs
print("Environment Ready")

  • Load the bag file with data and realsense pipeline.

# Setup RealSense pipeline and configuration:
pipe = rs.pipeline()
cfg = rs.config()
cfg.enable_device_from_file("object_detection.bag")
profile = pipe.start(cfg)

# Skip 5 first frames to give the Auto-Exposure time to adjust
for x in range(5):
    pipe.wait_for_frames()

# Store next frameset for later processing:
frameset = pipe.wait_for_frames()
color_frame = frameset.get_color_frame()
depth_frame = frameset.get_depth_frame()

# Cleanup:
pipe.stop()
print("Frames Captured")

  • Display the loaded image in rgb color.

# Display color frame:
color = np.asanyarray(color_frame.get_data())
plt.rcParams["axes.grid"] = False
plt.rcParams['figure.figsize'] = [12, 6]
plt.imshow(color)

  • Display the loaded image in colorized depth.

  • Display an image by comparing an RGB color image to a color depth image.

  • Load the model to use SSD-MobileNet and display the detected class.

  • Use the depth to guess the size of the detected object.

  • Measure the distance of the detected object.